Viewshed Validation

Comparison of MICRODEM LOS/Viewshed with other software, including ESRI.  They produce very similar results, and MICRODEM is at least as good as the others.  Randy Swanson was with the Topographic Engineering Center (now Army Geospatial Center, AGC), US Army Corps of Engineers and did two studies in 2003.

To validate viewsheds, you need to compare your results with an accepted program using the same data set and parameters.  When doing the comparison, you must be careful about distorting a viewshed when you take it into another program, or through raster to vector conversions.  Since both MICRODEM and other software (e.g. ArcGIS) do viewsheds as raster computations, the comparison should be done in the raster realm.  The difference will be most severe as the range increases, because many of the shortcuts for speed do not adequately consider curvature of the earth.  In particular, if you have a new algorithm, you must compare it carefully to previous algorithms; if there is a difference, it is most likely the new one.

Two steps to validation:

  1. Insure the algorithm works correctly, by comparing to other analyses.
  2. Compare to ground truth data, if you have it.

MICRODEM has a number of tools for this:

  1. Multiple DEM profiles to see the effect of different DEMs.
  2. Line of sight/Topographic Profiles which has a number of options. (LOS Right click options)

Viewshed/weapons fans references


Last revision 5/29/2015